Metraipolis › archive › dc_8d21056cf56d

When Attention Does Not Explain the Peak: Temporal Reference vs. Forecast Output in Attention-Based Time-Series Forecasting — arXiv:2610.07080

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.07080
License (as archived)
CC0-1.0
Archived
2026-10-07T11:40:51.177Z
SHA-256 of the archived text
b7f6e59fa620214b48f8e267145246f916e00f9d53fc09ce1fa475b0938d5b95
Stable URL of this page
https://metraipolis.com/documento/dc_8d21056cf56d

How to cite

When Attention Does Not Explain the Peak: Temporal Reference vs. Forecast Output in Attention-Based Time-Series Forecasting — arXiv:2610.07080 — arXiv (en), https://arxiv.org/abs/2610.07080 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8d21056cf56d (Metraipolis archive dc_8d21056cf56d, sha256 b7f6e59fa620214b48f8e267145246f916e00f9d53fc09ce1fa475b0938d5b95)
BibTeX
@misc{dc8d21056cf56d,
  title = {When Attention Does Not Explain the Peak: Temporal Reference vs. Forecast Output in Attention-Based Time-Series Forecasting — arXiv:2610.07080},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.07080},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8d21056cf56d; sha256 b7f6e59fa620214b48f8e267145246f916e00f9d53fc09ce1fa475b0938d5b95},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.07080. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Attention maps are often interpreted as evidence of what a forecasting model uses when making predictions. In our load-forecasting model, a CLS representation of historical demand queries 24 future exogenous horizon tokens through cross-attention, inviting a temporal interpretation in which highly attended horizons may appear to explain forecast peak timing. We test this interpretation using a horizon-level attention descriptor, $\Psi_{\mathrm{out}}$. Across 31 day-aligned windows of the Panama load dataset, the forecast achieves a median peak-time error of 0 h and a 51.6% exact-match rate, whereas the argmax of $\Psi_{\mathrm{out}}$ has a median error of 5 h and 0% exact match. The forecast peak is closer to the observed peak in 27 of 31 windows. This dissociation is not merely an argmax artifact: within $\pm1$ h, attention reaches only $1.16\times$, $1.11\times$, and $1.14\times$ the uniform baseline around observed, predicted, and weekly-naive peaks, respectively, indicating weak and non-selective concentration. Yet the attention profile is structured, with cross-window consistency of 0.83. Replacing 12 future weather features with their training-set means makes the profile nearly uniform, showing sensitivity to future weather variation rather than fixed horizon position alone. The dissociation is also reproduced across three random-seed runs. These results show that structured, input-sensitive, and reproducible horizon-level cross-attention need not provide a valid peak-selective explanation of forecast behavior. The observed behavior is instead consistent with an internal horizon-reference role for integrating future exogenous information, although this functional role is not causally established.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.07080 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T11:19:24.727Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).